Palliative care health professionals’ experiences of caring for patients with advance care directives

Advance care planning is an ongoing process that encompasses the exploration and documentation of a person’s preferences for future medical interventions and care, as well as ensuring the patient’s understanding of their condition.¹ These conversations need to be tailored to the patient’s circumstances, needs, wishes and concerns, and be facilitated by a health professional who has an understanding of the patient’s psychosocial circumstances and medical condition.

Optimal and effective advance care planning involves several key components: conversation about, and consideration of, patient values and wishes surrounding end-of-life and life-sustaining treatment;² communication about, and documentation of, patient values and goals for treatment (including continuing and withdrawal of active treatment) in an advance care directive (ACD);³ presentation of information specific to patient preferences around unacceptable health and/or functional states and locations for care and death;⁴ and nomination, where possible, of a substitute decision maker for situations where the patient is unable to actively participate in care decisions.^2,5 ACDs may be completed by healthy individuals in anticipation of a future time of physical deterioration or diminished decision-making capacity. Advance care planning may also be undertaken by people who have a life-limiting illness or when deterioration is imminent.⁶ The recent systematic review by Lovell and Yates explored the contextual factors influencing the uptake of advance care planning and development of ACDs.⁷ However, despite the reported benefits, our knowledge of the use of ACDs in the clinical setting is limited, particularly within specialist palliative care services.⁶ Specifically, little is known about how ACDs are implemented in palliative care or how advance care plans affect palliative care health professionals’ experiences of caring for people at the end of life, especially in the Australian health context. The aim of the present study was to explore service providers’ perceptions and experiences of advance care planning and the use of ACDs in Australian palliative care services.

A survey of Australian palliative care services was conducted using a self-report online questionnaire, consisting of closed and open-ended questions. This paper reports data from the latter. The directors of palliative care services involved in the Palliative Care Outcomes Collaboration (PCOC)⁸ were asked to nominate two health professionals who worked as clinicians within the service to complete the survey. Further details of the survey and results from the closed questions are provided elsewhere.⁶ Consent was implied by completion of the online questionnaire. Ethics approval was obtained from the University of Wollongong/Illawarra Shoalhaven Local Health District Human Research Ethics Committee (Approval no. HE12/471).

The three open-ended questions reported here asked participants about: (1) completing ACDs with patients from culturally and linguistically diverse backgrounds; (2) situations where ACDs were not followed; and (3) health professionals’ experiences of caring for patients with an ACD. The questions were formulated and modified by the research team to gain a broad understanding of advance care planning and the use of ACDs within the palliative care setting. Questions were reviewed by advance care planning and quality improvement facilitators from the Respecting Patient Choices and The Palliative Care Outcomes Collaboration programs^8,9 to ensure they were understandable and applicable to palliative care services.

The responses were entered into NVivo (Version 10, QSR International Pty Ltd, Melbourne, Australia), whereupon thematic analysis was used to explore and categorise the data. The principle of constant comparison guided an inductive, iterative process of data analysis.¹⁰ Data were coded collaboratively by authors (RS, CJ), using a matrix that underwent several modifications until the final framework of themes was developed. Once thematic analysis was complete, the data were split into groups by service provider type to determine areas of discrepancy and consistency across the coding framework. Data analysis and interpretation undertaken by the authors RS and CJ was verified by the authors MS, MM and WS and reconciled where necessary.^11,12

Sixty participants responded to at least one of the open-ended questions. This was 82% of health professionals who completed the full questionnaire.

The characteristics of participants are provided in Table 1. Most were female nurses, 40 years of age or older and from metropolitan services. Participants had spent a mean of 12 years working in palliative care. Less than half had received training in advance care planning and one-third reported completing an advance care planning discussion with a patient over the course of their career. Participants were similar in profession, age, service provider type and location of service to those who did not provide comments.

Table 1.  Participant characteristics (n = 60)
Note, data may not add to 100% due to rounding. CI, confidence interval

The categories identified were grouped into four key themes: the ACD; the process of developing ACDs; the process of using ACDs; and the consequences of having ACDs.

Health professionals discussed advance care planning in several formats, including formal ACDs, informal advisory documentation (usually documented within a patient’s medical record) and the appointment of medical guardians (Table 2). There was consensus that plans were useful only when they were documented, comprehensive, relevant and available. Inadequate levels of documentation were perceived to limit the usefulness of ACDs.

Table 2.  Textual examples of participants’ perspectives concerning advance care directives
N, nursing participants; M, medical participants; S, social worker participants

Consultant palliative care service participants were concerned that ACDs were of limited value if they were not explicit about medical situations encountered by the patients. Another participant noted that medical issues were sometimes poorly understood by patients and families and, unless ACDs were prepared with some medical input, they were often of little use.

Problems were reported when ACDs were prepared by people with little knowledge of the legal issues associated with the preparation and use of ACDs. This was a particular concern when advance care planning was left to junior medical staff who lacked the necessary skills and legal knowledge.

Another problem was the availability of ACDs, with several cases cited where patients were treated contrary to the patient’s and family’s wishes because patients were transferred to medical facilities without the appropriate original documentation. In these instances, active treatment was reportedly instituted until a documented ACD was made available to the medical team.

In the experience of participants, the process of advance care planning was not routinely undertaken for patients, either in the community or in-patient setting. Patients admitted to services at the end of life frequently did not have an ACD and, in some instances, participants reported that advance care planning was not undertaken during an admission either (Table 3). Some health professionals described health settings where relevant policies were not in place and advance care planning was not routinely conducted.

Table 3.  Textual examples informing the theme about developing advance care directives
N, nursing participants; M, medical participants; S, social worker participants; ACDs, advanced care directives

Numerous participants identified advance care planning as an iterative process that needs to respond to changes in patients’ circumstances and condition. Several participants identified that a patient’s wishes may change as they are faced with circumstances that, up until that time, had been hypothetical.

Communication was central to the development of ACDs. The development of an ACD was perceived to support conversations within families and between the patient, family and healthcare team, and made communication about end-of-life care easier. Participants reflected upon the need for ongoing conversations to make sure ACDs were current and appropriate for changing circumstances, and to negotiate suitable goals of care. Even though such conversations were time-consuming, they were perceived to be important to build trust and to ensure that ACDs remained relevant.

Participants recognised their legal and moral responsibilities to adhere to patients’ wishes ‘unless there is a clear and compelling reason not to do so’ (M6; Table 4), but situations were described where this did not occur (Table 4). Health professionals from all occupations recognised that the use of ACDs was limited, for varying reasons: the documentation was unavailable, incomplete, ignored, not discussed or no longer appropriate to the patient’s circumstances. Nursing staff also commented on occasions when they had to challenge the treating team about not respecting an ACD.

Table 4.  Textual examples informing the theme about the process of using advance care directive
N, nursing participants; M, medical participants; S, social worker participants; GP, general practitioner; CVA, cerebral vascular accident; abx, antibiotics; NG, nasogastric

Participants discussed several factors that were perceived to influence the use of ACDs. First, the lack of health service, state or national policy regarding ACDs was viewed as a confounding factor in their implementation. Several participants reported that their services did not recognise ACDs or have policies in place for their use.

Second, participants reported that ACDs were not always actively looked for when patients presented at the hospital. Cases were reported where a clinician was informed (either by family members or palliative care health professionals) that a patient had an ACD but active management was pursued until the ACD was produced.

Third, the usefulness of an ACD was perceived by participants to be influenced by its relevance to the patients’ circumstances. Medical issues are reportedly often poorly understood by patients and families and, hence, the usefulness and relevance of information contained in ACDs is variable. Changes in a patient’s condition that were not addressed in the ACD were also believed to make the plan irrelevant.

Fourth, participants reported that compliance with a patient’s wishes was more difficult when there was disagreement between interested parties (e.g. various family members or, in some cases, treating doctors). These situations were described by participants across all types of services, particularly in circumstances involving poor documentation.

Health professionals observed that advance care planning was ‘empowering for the patient and family’ (S2; Table 5). Patients who had an ACD were perceived to be more proactive and in control of their care. Some participants reported that having clear documentation of a patient’s expressed wishes helped health professionals to advocate on the patient’s behalf. Participants reported that both families and health professionals were more confident about making treatment decisions when they had insight into ‘…what the patient really wants’ (N21). Family members were perceived to be more comfortable and accepting, knowing that the decisions made were what the patient wanted or reflected the patient’s values and priorities. Many participants mentioned a sense of relief when treatment decisions reflected the wishes and priorities of their patients. Several participants reported a sense of satisfaction when they were able to provide the care patients desired. Providing end-of-life care in such a context was described as a rewarding process.

Table 5.  Textual examples informing the theme ‘Consequences of having an advance care directive’
N, nursing participants; M, medical participants; S, social worker participants

Participants also reported that advance care planning made communication about end-of-life care easier with both the patient and the family, knowing that the person had already thought about these issues. However, one doctor stated that conversations about end-of-life care were still difficult, even when previous conversations about advance care planning had been documented. Patients were not always perceived to be in control and, in some situations, reportedly received treatment that was contrary to their expressed wishes.

This paper reports the advance care planning experiences of palliative care health professionals and highlights the role of ACDs in facilitating communication, decision making and providing care for dying patients. ACDs promoted greater understanding between family, friends and staff, and reduced distrust and conflict. However, the use and applicability of ACDs was dependent on the availability, relevance and quality of the documentation.

Participants primarily reported positive experiences for health professionals and patients when advance care planning was undertaken, and both the development and use of ACDs in practice was perceived to be rewarding.

The present study identified several barriers to the advance care planning process and use of ACDs in practice that are similar to previous research.⁷ The need for ACDs to be accessible and available to all health professionals involved in the patient’s care and across health jurisdictions has consistently highlighted concerns about ownership of the ACD and sharing of patient information in a fragmented health system.⁷ To be effective, formal ACDs need to be available to the patient and their family, as well as to relevant health providers when health care decisions are made. Particular difficulties in accessing ACDs have been observed when health professionals become involved in the care late in an illness or in an emergency situation.¹³ In such situations, it is important for information to be immediately available, current and relevant to the patient’s circumstances, otherwise the default position is frequently to treat actively in the absence of a trustworthy ACD.^13,14

Concerns have also been raised about how to maintain an up-to-date ACD, because changing personal circumstances have been shown to result in changing priorities for care.¹⁵ Individuals may have different treatment preferences when they are healthy to those when they are ill and, as a consequence, choices about what treatment is desirable and acceptable may change over the course of their illness.^16,17 Although GPs are well positioned to initiate ACDs due to their ongoing relationship with their patients, how such information is consistently shared across health care settings remains problematic. Numerous approaches to sharing ACDs have been investigated both in Australia and internationally, but there has been limited universal uptake due to a lack of systematic processes for embedding them into routine practice at an institutional level and across health jurisdictions.¹⁸ Furthermore, advance care planning needs to be normalised in all health settings. To date, the greatest hope for embedding ACDs across services and jurisdictions is the electronic medical record, which is an important facilitator of advance care planning.¹⁹ However, in Australia, universal E-health records are slow to be implemented and hampered by limited consumer use.²⁰

Participants in the present study identified the importance of communication in advance care planning; indeed, that advance care planning is an iterative process that is predicated on ongoing discussions about the patient’s preferences for treatment. Our findings support the notion that having an ACD in place, as well as the preparation of an ACD, are access points for ongoing conversations about what treatment and care is preferred and is appropriate. Such discourse between health professionals, the patient and, where appropriate, the family is reflective of a patient-centred approach that can be responsive to the changing health and differing decision-making styles of individual patients.²¹ Communication at this level helps build trust between health professionals, patients and their families, as well as an understanding of the care preferences of the patient.¹³ Although palliative care is built on an assumption of open and honest communication, staff in this study reported that they find it difficult to discuss end-of-life issues. Thus, advance care planning and the preparation and use of ACDs provide an opening for the discussion of issues of significant importance to the patient and their loved ones.

The main limitations of the present study relate to the generalisability and representativeness of the data. Although 60 health professionals provided open comments about their experiences, we do not know from which palliative care services these responses came. However, services participating in PCOC are representative of Australian palliative care services generally.²² Furthermore, service managers were asked to distribute the survey to well-informed staff members. Qualitative research aims to include people from a variety of backgrounds who are willing and able to discuss the phenomenon in depth, focusing on their experiences of the issues under discussion. Greater numbers or representativeness of the sample ‘will not necessarily result in a better understanding of the topic’.²³

To increase the reliability of these findings, an investigation of doctors’ perceptions of the clarity and applicability of ACDs when developed in various settings would be helpful. Further research is also warranted to provide insights into patient experiences of advance care planning in palliative care.

Nevertheless, our research has several implications for advance care planning in the clinical setting. First, ACDs must be clear, comprehensive, medically relevant, transportable documents. Second, clear communication channels are necessary for the development and application of ACDs. Because ACDs are individualised, their development and maintenance is highly dependent on an intimate understanding of the patient and his/her desires. Third, having an ACD is not an end in itself. Discussions about preferences for care must be ongoing between health professionals, patients and their families. Any changes in patients’ wishes need to be documented and effectively communicated to the healthcare team and across health jurisdictions.

The authors declare that there are no competing interests.